(Slider with Hydroplus®, Mansfield Boston Scientific Corp., Watertown, MA). The catheter was 
constructed with a single polyethylene balloon, 2.0 mm in diameter and 2.0 cm in length. The 
Hydroplus® coating consists of a hydrophilic polyacrylic acid polymer, crosslinked via an isocyanate onto 
the balloon to form an ultra-high molecular weight hydrogel with tight adherence to the balloon surface. 
The thickness of the hydrogel coating when dry measures between 3-5 |im; upon exposure to an aqueous 
environment, the coating swells to 2 - 3 times its initially dry thickness. In order to apply DNA to the 
catheter, the balloon was inflated to 4 atm, following which 20 (ll of DNA solution were pipetted and 
distributed onto the balloon surface using a sterile pipette tip. After the balloon's hydrogel polymer was 
covered with a homogeneous film of DNA solution, the hydrogel was dried with warm air. The above 
procedure was then repeated, resulting in a total of 40 [ll of DNA solution applied to the balloon. 
For percutaneous application, luciferase DNA concentration was 3.27 |ig/|il. DNA was 
dissolved in TE buffer (10 mM Tris, 1 mM EDTA). 
(Attempts were made to apply DNA solution to standard uncoated balloons as well. The 
hydrophobic surface of the polyethylene balloon, however, made it impossible to cover the balloon with a 
film of DNA solution.) 
To determine the total amount of DNA which was successfully absorbed onto the balloon 
surface, 5 hydrogel balloons were coated with 40 |il DNA (2 fig DNA/|il) containing a small amount of 
35s-labeled luciferase plasmid. A random primed DNA labeling kit (United States Biochemical, 
Cleveland, OH) was used for the labeling reaction and unincorporated nucleotides were removed by 
ethanol precipitation. After the coating procedure, the catheter tip was placed in 0.5 ml water for 15 min at 
room temperature, and 1.0 ml gel solubilizer (Solvable, New England Nuclear, Boston, MA) for 3 hours 
at 50°C to dissolve the gel before the scintillation fluid was added. The amount of DNA on the balloon was 
calculated from the quotient: [counts per minute (cpm) in a scintillation vial containing the balloon]/[cpm in 
a vial containing 40 |il of the same lot of labeled DNA (80 fig)]. Scintillation counts were corrected for 
quench and chemiluminescence. 
Percutaneous gene transfer experiments with the luciferase gene were performed in 13 rabbits 
using a catheter with a balloon to which a 20 ftm hydrogel coating had been applied and which was 
advanced through a 5 F teflon sheath. The balloon was advanced beyond the distal tip of the sheath, coated 
with 130 fig luciferase DNA, and pulled back into the sheath to protect the balloon from subsequent 
contact with blood. The sheath and the angioplasty catheter were then introduced via the right carotid artery 
and advanced to the left common iliac artery under fluoroscopic control. The balloon catheter was 
advanced 3 cm further (beyond the distal sheath tip) into the external iliac artery and inflated there for 1 or 
5 min. Following balloon deflation, the catheter system was removed. In 10 animals, the transfected 
external iliac artery as well as the contralateral control artery were removed 3 days later, weighed, and 
assayed for luciferase activity. In 3 additional animals, which had been transfected for 5 min only, the 
arteries were excised 14 days after gene transfer. In these 3 animals we also removed the left femoral 
artery to check for luciferase expression directly downstream of the transfected segment 
After coating hydrogel balloons with 40 (ll of DNA solution (containing 80 (Ig of radiolabeled 
DNA), and drying the gel, the magnitude of DNA retained on the hydrogel balloon was determined by 
comparing the amount of radioactivity on the balloons to the amount of radioactivity in 40 |il of the original 
radiolabeled DNA solution. Scintillation counting revealed that 97 ± 2% (n=5) of the radioactively labeled 
DNA remained on the hydrogel coated balloon, corresponding to 78 ± 1.5 jig of luciferase DNA. 
In a small series of 3 animals, the gene for nuclear specific (3-galactosidase (nls [3-gal) was 
used in lieu of the luciferase gene as a marker gene. The (3-gal assay allows the histological identification 
of positively transfected cells. The typical nuclear staining pattern, which is specific for transfected cells, 
was detected in cells found predominantly in the subintimal and occasionally in the intimal layers. The total 
number of intimal and medial cells transfected, however, was nevertheless low (< 0.1%). 
Luciferase expression was detected in all 10 (100%) percutaneously transfected arteries excised 
after 3 days, whether inflated for 5 min (386 ± 299 TLU, n=5) or 1 min (1 13 ± 59 TLU, n=5). 
Three additional animals, in which balloons were inflated for 5 min only, were sacrificed after 
14 days. Individual luciferase expression was 152, 6, and 16 TLU, respectively (mean = 58 ± 47 TLU). 
In this series we also measured luciferase in the adjacent femoral artery, which was not inflated. Luciferase 
expression in all these arteries was undistinguishable from background activity (mean 0.04 ± 0.29 TLU). 
The findings reported here demonstrate that endoluminal vascular gene transfer can be achieved 
successfully and consistently with pure DNA applied to a standard angioplasty catheter balloon coated with 
hydrogel polymer. The hydrogel provides the absorbable medium to which one may apply a solution of 
pure DNA. Drying of the gel results in a layer of concentrated DNA which is then transferred to the arterial 
[320] 
Recombinant DNA Research, Volume 20 
